- Hewlett-Packard Reference Manual

Chromatographic Troubleshooting
Baseline symptoms
206
Spiking
Spikes are isolated baseline disturbances, usually as sudden (and large)
upscale movements. If accompanied by noise, the noise problem should be
solved first, since spiking may disappear at the same time.
1. Spikes appear whenever the chart is running:
The cause is almost always electronic in origin: Loose connections
are likely. Check signal cable connections at the detector and
controller ends.
A dirty slidewire on a recorder may cause this; see the
maintenance section of your recorder manual.
Loose or dirty contacts between printed circuit boards and their
connectors may be responsible. Read appropriate sections
regarding servicing boards and connectors for the HP 5890.
2. Spikes appear on chromatograms but not when the recorder is
isolated (no input signal):
This indicates a detector problem: In extremely dirty FID
collectors, particles may break away, causing disturbances in the
most sensitive area of the detector. This appears on the chart as a
spike.
A less common cause is column packing particles being blown into
the detector: A loose or inadequate glass wool plug at the detector
end of the column may fail. As particles are blown through the
detector, the disturbance generates a spike. This may happen with
any detector, but FIDs and NPDs are particularly susceptible
because of the narrow bore of the jet.
The bottom of the jet in FIDs and NPDs extends into the end of a
packed column. If the bottom touches the glass wool plug at the
end of the packing, spikes are produced.